agent-school
Investigates a problem area in the codebase and generates a new tessl tile (rules, docs, skills) to teach agents how to handle it correctly. Use when agents keep making the same mistakes around a library, design pattern, or convention.
84
76%
Does it follow best practices?
Impact
100%
1.36xAverage score across 3 eval scenarios
Passed
No known issues
Optimize this skill with Tessl
npx tessl skill review --optimize ./agent-school/SKILL.mdQuality
Discovery
75%Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.
This description effectively communicates its purpose and includes explicit 'Use when' guidance, making it strong on completeness and distinctiveness. However, it could benefit from more specific concrete actions and additional natural trigger terms that users might employ when they need this skill.
Suggestions
Add more specific concrete actions like 'analyzes error patterns', 'extracts problematic code examples', 'generates rule definitions with examples'
Expand trigger terms to include natural phrases like 'recurring errors', 'agent keeps failing', 'teach the agent', 'create a rule for this'
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Names the domain (tessl tiles - rules, docs, skills) and describes the general action (investigates problem area, generates tiles), but lacks specific concrete actions like 'analyzes error patterns', 'extracts code examples', or 'creates rule definitions'. | 2 / 3 |
Completeness | Clearly answers both what ('Investigates a problem area and generates a new tessl tile') and when ('Use when agents keep making the same mistakes around a library, design pattern, or convention') with explicit trigger guidance. | 3 / 3 |
Trigger Term Quality | Includes some relevant terms like 'mistakes', 'library', 'design pattern', 'convention', but misses natural variations users might say like 'recurring errors', 'agent keeps failing', 'teach Claude', 'create a rule', or 'document this pattern'. | 2 / 3 |
Distinctiveness Conflict Risk | Has a clear niche focused specifically on creating tessl tiles for agent learning from repeated mistakes - unlikely to conflict with general documentation or code analysis skills due to the specific 'tessl tile' and 'agent mistakes' framing. | 3 / 3 |
Total | 10 / 12 Passed |
Implementation
77%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is a well-structured, actionable skill with excellent workflow clarity and concrete examples. The five-phase approach with explicit user checkpoints is a strength. However, the skill is somewhat verbose and could be more concise by assuming Claude's knowledge of basic concepts, and the lengthy content could benefit from progressive disclosure to separate reference material.
Suggestions
Remove explanatory content about what tiles contain in the introduction - Claude can infer this from the structure
Extract the tile.json schema examples and directory structure templates into a separate REFERENCE.md file, linking to it from the main skill
Condense Phase 2 sub-sections by removing obvious instructions (e.g., 'use Grep and Glob to locate them' is implied)
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is reasonably efficient but includes some redundant explanations (e.g., explaining what tiles contain when Claude likely knows this). The tables and structured content are helpful, but some sections could be tightened. | 2 / 3 |
Actionability | Provides concrete, executable guidance throughout: specific file paths, exact JSON structures for tile.json, directory structures, and specific commands like 'tessl install' and 'tessl status'. Examples are copy-paste ready. | 3 / 3 |
Workflow Clarity | Excellent multi-phase workflow with clear sequencing (5 phases), explicit checkpoints ('Wait for user approval before proceeding'), and verification steps in Phase 5. Each phase has clear sub-steps and the process includes feedback loops (confirm with user, verify installation). | 3 / 3 |
Progressive Disclosure | Content is well-organized with clear headers and phases, but it's a monolithic document (~200 lines) that could benefit from splitting detailed reference material (like tile.json schemas, directory structures) into separate files. No external references are provided. | 2 / 3 |
Total | 10 / 12 Passed |
Validation
100%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 11 / 11 Passed
Validation for skill structure
No warnings or errors.
- Repository
- jamesmoss/skills
- Commit
df78c2e
- Reviewed
Table of Contents
Is this your skill?
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.